Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Cmos inverter/nor/nand gates dimensions

Status
Not open for further replies.

AMSA84

Advanced Member level 2
Joined
Aug 24, 2010
Messages
577
Helped
8
Reputation
16
Reaction score
8
Trophy points
1,298
Location
Iberian Peninsula
Activity points
6,178
Hi guys,

I'd like to know if someone can tell me which are the criteria do design a NOR gate in terms of MOSFETs dimensions. The tech is 130nm.

Should I use the minimum transistor sizes in all mosfets?

Thanks in advance.
 

For equal rise/fall times You have to provide the same resistances for group of pfets and nfets of each gate. In first approximation the Wp/Wn =Kn/Kp for inverters. For nand gates the parallel connection of pfets should give you the same resistance as serial connection of nfets, while opposite for nor. The length should be set a minimum one (it's not true for radiation hardness designs) and width depends to load capacitance and maximum freuqency of your design.
 

Thanks.

There are any expressions where we can estimate in a first order analysis before simulate?
 

The rise/fall time could be calculate from constant current charging/discharging load capacitance. The current value is VDD·Ron, where Ron is resistance of each transistor and is equal to:
\[R_{on}=\frac{1}{K W/L (V_{dd}-V_{th})}\]
The output voltage could be simply calculated as exponential function with time constant equal to Ron·Cload (small signall analysis) or by using slew rate equation SR=Vdd/(Ron·Cload) [V/s] which also gives You a time of linear charging load capacitance from ground to Vdd.

In fact I'm not sure which one gives better results
 

Yes, yes. Of course Vdd/Ron. I solved Too much nodal equations with conductances instead of resistances ;-)
 

Thanks for the reply. I understand.

Allow me to ask another thing.

When we're designing a non overlap circuit, can we use different number of delay elements in the feedback path?

For example, on the 1st clock 2 inverters and in the 2nd clock 4 inverters?

Regarding the non overlap circuit, why we can only use a even number of delay lines? (2, 4, 6, etc).
 

Non-overlapping clock pulse generator

When we're designing a non overlap circuit, can we use different number of delay elements in the feedback path?
For example, on the 1st clock 2 inverters and in the 2nd clock 4 inverters?
Sure, you can do this. Creates some asymmetry re. clock pulse & non-overlapping time.

Regarding the non overlap circuit, why we can only use a even number of delay lines? (2, 4, 6, etc).
You always need an uneven (odd) number of inverting stages (180° phase shift) in total between clock input and feedback input, s. this figure: **broken link removed**
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top